similar to: e1071 SVM: Cross-validation error confusion matrix

Hi, I ran two svm models in R e1071 package: the first without cross-validation and the second with 10-fold cross-validation. I used the following syntax: #Model 1: Without cross-validation: > svm.model <- svm(Response ~ ., data=data.df, type="C-classification", > kernel="linear", cost=1) > predict <- fitted(svm.model) > cm <- table(predict,

Hi all, How can one evaluate NAs in a numeric dataframe column? For example, I have a dataframe (demo) with a column of numbers and several NAs. If I write demo.df >= 10, numerals will return TRUE or FALSE, but if the value is "NA", "NA" is returned. But if I write demo.df == "NA", it returns as "NA" also. I know that I can remove NAs, but would like

hi, sorry for the potentially stupid question. I'm a little stuck. when trying to run the stupid update manager i need for my satnav, which - obviously - is windows only, nothing happens. install worked, the icon's on my desktop, it has been made executable, but after a bit of harddrive action nothing happens. terminal says: Code: mzwo at mzwo-laptop:~$ env

Dear r-helpers, I have a question about multiple testing. Here an example that puzzles me: All matrixes and contrast vectors are presented in treatment contrasts. 1. example: library(multcomp) n<-60; sigma<-20 # n = sample size per group # sigma standard deviation of the residuals cov1 <- matrix(c(3/4,-1/2,-1/2,-1/2,1,0,-1/2,0,1), nrow = 3, ncol=3, byrow=TRUE, dimnames =

I found: max.col(matrix(c(1,3,2),nrow=1)) Is there a more concise/elegant way? Thanks, ----- TO GET MORE DETAILS CLICK HERE -- View this message in context: http://r.789695.n4.nabble.com/finding-index-of-maximum-value-in-vector-tp4651663.html Sent from the R help mailing list archive at Nabble.com.

I'm trying to make a set of contour plots of bivariate kernel density estimates, showing three such plots overlaid, similar to this plot http://euclid.psych.yorku.ca/SCS/Private/Test/ridge-boot2.pdf except that I would like to have the contours *filled* (using transparent colors). To make this reproducible, I've saved the results of KernSmooth::bkde2D() in the following file:

Hello there, I would like to simulate X --Normal (20, 5) Y-- Normal (40, 10) and the correlation between X and Y is 0.6. How do I do it in R? Thank you very much ----- TO GET MORE DETAILS CLICK HERE -- View this message in context: http://r.789695.n4.nabble.com/How-to-simulate-correlated-data-tp4651667.html Sent from the R help mailing list archive at

Hi I've suddenly lost access to some permanent cross-server shares. I think this was following the Nov 5 new kernel, rather than the Centos version upgrade. I don't see any difference to man mount.nfs I have:- [root at hs6 ~]# mount hs8:/data /sysback/hs8-data -v mount: no type was given - I'll assume nfs because of the colon mount: trying 172.26.0.8 prog 100003 vers 3 prot tcp port

Is that possible to create superscript text on the graph legend, for example to put "cm2" (centimeter square) on the legend. Please show me how to do it. Thanks. -- Agus Susanto [[alternative HTML version deleted]]

Hi, I am in the process of linking a C/C++ application to the R.dll directly. I have obtained the R source code and compiled it successfully. I have also successfully linked the R.dll directly into our application and have made calls successfully into the R.dll that are included in the sample rtest.c and in the "Writing R Extensions - The R API". The R functionality that we

I am trying to fit a nonlinear regression to infiltration data in order to determine saturated hydraulic conductivity and matric pressure. The original equation can be found in Bagarello et al. 2004 SSSAJ (green-ampt equation for falling head including gravity). I am also VERY new to R and to nonlinear regressions. I have searched the posts, but am still unable to determine why my data come up

I thought this had been fixed with newer kernels? We have an old Maxtor Snap Server, an 80 GB NAS unit, containing some of our archives. Both of our CentOS servers have it permanently mounted by NFS for convenient access. This weekend, I discovered remotely that, although all systems except the VRTSralus remote agents for Veritas (now Symantec) Backup Exec were working just fine, both servers

Well, one way to do it is xyplot(y~time|id, data=dat,type='l', panel=function(x,y,subscripts,...){ panel.xyplot(x,y,subscripts,...) panel.abline(v=dat[subscripts,"cm1"]) panel.abline(v=dat[subscripts,"cm2"]) } ) Since I don't know what the dataframe 'mergeData' is I have used 'dat'. Best regards Frede Aakmann

Consider this data frame: dat<-data.frame(id=gl(3,5),time=rep(1:5,3),cm1=rep(c(2,3,4),each=5),cm2=rep(c(2.5,3.5,4.5),each=5),y=rnorm(15)) > dat id time cm1 cm2 y 1 1 1 2 2.5 -1.0824549 2 1 2 2 2.5 -0.7784834 3 1 3 2 2.5 -1.7783560 4 1 4 2 2.5 0.5056637 5 1 5 2 2.5 -0.1967505 6 2 1 3 3.5 1.1643136 7 2 2 3 3.5 -1.4323765 8 2

Hi, I have a question about the parameter C (cost) in svm function in e1071. I thought larger C is prone to overfitting than smaller C, and hence leads to more support vectors. However, using the Wisconsin breast cancer example on the link: http://planatscher.net/svmtut/svmtut.html I found that the largest cost have fewest support vectors, which is contrary to what I think. please see the scripts

Hi all, is it possible to concatenate expressions and basic text when for instance labeling axis of a plot? I would like to see something like the concatenation of expression(C[0]) and "for case 1" on my x axis. Obviously a plot(x, y, xlab=paste(expression(C[0])," in case1")) will not work. Thank you in advance, ----- TO GET MORE DETAILS CLICK HERE --

I'm working with large data frames and running out of memory. I hope some of you may be able to suggest a more efficient approach. I have grid/lattice data representing a time series of 1 m2 quadrats in a grassland: Each 1 cm2 cell or pixel contains one ecological state (ie grass or bare ground). The goal is to calculate, for each cell, the transition probabilities to all available

Hi I get this on every scan report:- LibClamAV Warning: *********************************************************** LibClamAV Warning: *** This version of the ClamAV engine is outdated. *** LibClamAV Warning: *** DON'T PANIC! Read http://www.clamav.net/support/faq *** LibClamAV Warning: *********************************************************** ----------- SCAN SUMMARY

I'm just wondering if anyone else here is using this? The majority of jobs on our CentOS5 server "fail" by "Communications failure" right at the end, apparently on the very last file which, however, is recoverable, as is the whole archive. In more than half these failures, the Linux agent has stopped and, in about half of those, crashes leaving a backtrace. I had similar

Hi, I'm trying to figure out how anova works in R by translating the examples in Sokal And Rohlf's (1995 3rd edition) Biometry. I've hit a snag with planned comparisons, their box 9.4 and section 9.6. It's a basic anova design: treatment <- factor(rep(c("control", "glucose", "fructose", "gluc+fruct",